The new ice age
In Elmsford, N.Y., superconducting microelectronics firm HYPRES has developed a chip solution that addresses major radio frequency (RF) challenges in each of these arenas while causing industry participants to rethink wireless system designs and business models
At the heart of HYPRES’ breakthrough is a superconducting niobium chip — housed in a cyrocooler at 4° Kelvin (-452° Fahrenheit) — that provides direct digitization of signals at an unprecedented range of frequencies.
“What we’re trying to do is replace the RF chain with a digital chip,” said Deep Gupta, vice president of research and development for HYPRES.
The company’s technology has captured the attention of the military, which views the HYPRES solution as a way to dramatically simplify the RF chain in its satellite systems while dramatically enhancing system performance. In a February test at the Joint SATCOM Engineering Center at Fort Monmouth, N.J., HYPRES demonstrated that its all-digital receiver (ADR) could digitize an X-Band satellite signal at 7.676 GHz — a task currently executed by a bank of 56 down converters.
Not only is the military’s current analog-to-digital system cumbersome, it introduces unwanted noise that degrades the satellite signal, said Rick Dunnegan, a telecommunications specialist with the Communications-Electronics Research, Development and Engineering Center who is leading on the project. Meanwhile, the HYPRES system promises to remove the need for the down converters and the low-noise amplifier (LNA) and, most important, the noise those components introduce into the system.
Ultimately, the efficiencies enabled by the HYPRES solution will improve the link with the satellite dish, saving valuable power across the satellite system that will translate into better data rates, Dunnegan said.
“The power that you save across that satellite can allow you to squeeze in a bunch of higher-order modulation schemes and, therefore, support a much higher throughput,” he said. “Power and bandwidth are directly proportional. If I double the amount of bandwidth, I will require twice as much power, or three more decibels.
“How much more video can we get to the soldier in the field? This is bigger bandwidth, so can we handle soldiers at an individual level, instead of just at a brigade or battalion level? Let your mind run with it, and you’ll be pretty amazed.”
KEEPING IT COOL
Key to the HYPRES solution is the superconducting niobium chip that requires temperatures approaching absolute zero (0° K) to function.
“As soon as these chips get to about nine Kelvin, they quit working,” said Richard Hitt, HYPRES CEO. “They’re just a hunk of metal — they don’t do anything.”
Maintaining such a cold temperature is cryocooling technology, which has matured considerably during the past decade, when it has been deployed increasingly in spacecrafts and demonstrated its reliability over the past few years, said Ted Nast, a fellow for Lockheed Martin.
During the military demonstration, HYPRES used a Sumitomo piston-compressor cryocooler that reaches the desired temperature range in two stages — from 300° K to 70° K, and then from 70° K to 4° K. Although the Sumitomo system is reliable and provides the desired performance, HYPRES officials wanted a cryocooler that would be more power efficient and smaller.
In the future, HYPRES plans to use a cryocooler built by Lockheed Martin that provides four stages of cooling and significantly better power efficiency in a form factor the size of a thermos. In addition, it should be more even more reliable and last longer, Gupta said.
“These pulse tubes are intrinsically robust because it’s gas moving through metal orifices in metal pipes,” Gupta said. “Gas moving through a solid metal pipe is much more robust than a piston moving up and down in a chamber because friction and other things means parts will wear out. If you have gas moving through a tube, what can go wrong?”
While used most in the relatively predictable environments of space, cryocooler operations should be just as reliable within the earth’s atmosphere, said Lockheed Martin’s Nast, who noted his company’s expects to deliver its new cryocooler to HYPRES in June.
“Definitely, the terrestrial environment is a tougher environment because in space, the temperature stability is pretty good — you don’t have vibrations or big temperature excursions like you do on the ground,” he said. “But there’s no indication that these space systems operated on the ground will have any limitations.”
Perhaps the greatest limitations will be man-made. Upon seeing the HYPRES solution, most systems engineers’ first instinct will be to use the technology to simply replace the components in the existing systems, Hitt said. But the presence of multistage cryocooling offers opportunities to do much more, he said.
“Things that work at ambient temperature will work better at controlled ambient temperature that is slightly cooled,” Hitt said. “All these processors that work at room temperature, if we could provide some amount of cooling and leave it close to a freezing temperature, it will work better.”
As evidence, Hitt noted the fact that researchers from IBM and Georgia Tech last year demonstrated that the performance of high-frequency silicon-germanium transistors could be improved by 40% — allowing them to operate at frequencies above 500 GHz — by placing the chips in cryogenic conditions.
With this in mind, Hitt said HYPRES is approaching companies in all aspects of the communication chain about the technology, asking them to consider how they would redesign their systems if cryogenic cooling was readily available. Gupta agreed that this is the proper approach but said he will not be surprised if some are resistant to change.
“If I was on the other side of the fence, maybe I would do the same thing,” Gupta said. “If somebody came in, showed me one demo and said, ‘Hey, I’m going to change your whole communications architecture,’ I would probably say, ‘Wait, let’s take out one box and put in your box. Let’s do that first before we talk about changing my architecture.’ That is our problem.”
Indeed, Dunnegan said he would question the notion of building systems that revolve around cryocooling technology, noting a historical precedent that indicated similar technology can be developed at room temperature.
“We had the cryogenics in there on the LNAs just to try to lower that noise. Efforts like this enabled us to get to a room-temperature LNA over time, and look at the money we saved in that venture — that was very significant,” Dunnegan said. “So what I see is that, rather than put more dependency on the cryogenics, we’ll most likely try to remove some of the functions that occur on the chip toward room temperature so that we can reduce the cryodependency.”
OTHER POSSIBILITIES
Cryogenically cooled or not, the HYPRES solution certainly is “groundbreaking” technology that has the potential to significantly improve communications, Dunnegan said. The Fort Monmouth test was particularly impressive because the performance of HYPRES’ system was comparable to that of the existing system, even though the solution has not been optimized and the demonstration dealt with an undersampling of the satellite signal.
While the next step for the military is to test whether HYPRES’ solution can perform similarly in the Ka satellite band, Dunnegan is quick to note that the technology will have significant effects outside the sector.
“We already have a pretty good grasp on things that we know of … but there are so many other things that I think we have no idea what we’re on the edge of right here,” he said.
In fact, the HYPRES chip did not just directly digitize the 7.676 GHZ X band signal, it digitized all signals in a 400 MHz swath in the band — the kind of frequency agility that industry experts have been anticipating to help enable SDR and, eventually, cognitive radio systems.
Although technology on the waveform-flexibility side of the SDR equation has matured, the frequency-agility side of SDR is “not really ready for prime time,” said John Chapin, chief technology officer for software radio vendor Vanu, which has a waveform-flexibility solution. Announcements from chip companies like Bitwave, HYPRES and TechnoConcepts indicating that frequency-agile RF transceivers could be commercially available soon are encouraging, Chapin said.
Because of its dependence on cryogenics, the HYPRES solution is not practical for handsets like the other two solutions. However, Chapin said HYPRES offers considerable promise for base stations because of its potential to receive very high bandwidth and its sensitivity — or spurious free dynamic range — that should be advantageous when a strong signal such as a TV transmitter is in the vicinity.
“Ordinarily, we put an analog filter into our systems, so we only hear the signals we want to hear. The promise of what HYPRES is talking about is a system that doesn’t require you to have that analog filter,” Chapin said. “You can have a wide open front end; they just digitize the whole spectrum, and there’s so much spurious free dynamic range that you can avoid getting swamped by that really strong signal … and still be able to hear the one you want to hear.”
Meanwhile, Hitt said HYPRES would like to work with a company like Vanu that provides the modulation technology — able to handle a variety of waveforms, from GSM to CDMA to 3G technologies — that complements his company’s frequency-agility technology.
“Marry those two together, and you essentially have a totally programmable system that ostensibly runs the gamut of current and future mobile wireless protocols and frequencies,” Hitt said.
A NEW TOWER MODEL?
If realized, this vision has the potential to alter the way wireless networks are constructed, Hitt said. Today, tower sites often support multiple antennas and base stations from a variety of carriers, each of which must pay for the design and deployment of the equipment.
By combining frequency agility and waveform flexibility in a single package, tower owners could serve the needs of multiple carriers with a single set of RF equipment, Hitt said. It’s a model that could provide new opportunities for tower owners while significantly lowering network deployment costs for carriers, he said.
“What if, instead of selling space on the tower, the guy who owns the tower sells digitized bandwidth?” Hitt asked. “So if I’m Sprint or someone else, I have a license to operate in a particular frequency band. But now, I don’t have to bring my own transceiver; I just have to buy from [a tower owner a] digitized signal in the band that I own, because you’re digitizing it for everybody on the tower.”
Such an arrangement also could be used to help make a nationwide public-safety network a reality, Hitt said.
“Normally, the public-safety guys have a hard time finding places to be or they have their own tower system,” Hitt said. “But what if they were just part of the digitized bandwidth that this newly equipped tower operator is now providing? They could pluck their bandwidth out of this digitized space, just like the commercial guys.
“The point is the tower becomes multifunctional between the commercial guys, the public-safety people, the FEMA people and whoever else is on there. And it’s done with almost literally one set of RF equipment. Today, it’s not that clean yet, but we’re very close to that.”
And HYPRES officials believe its technology is just scratching the surface of what should be an evolutionary path that should see regular performance improvement such as that experienced in the semiconductor industry according to Moore’s Law. Eventually, advances in micro-mechanical-electro-machines technology could allow a HYPRES niobium chip to be placed in handsets, but it will be many years before that can be considered, Hitt said.
Other possible future uses for the HYPRES system that could be realistic in the foreseeable future — after HYPRES is able to optimize the integration between its system and the cryocooler — include enhancing optical communications, enabling better mobile communications from vehicles and creating an ideal communications product for consumers.
“There is no reason to believe that you couldn’t have a compact version of our receiver sitting in your garage doing multimode communication,” Gupta said. “But we are not there yet — it’s a dream, not reality.” For now, HYPRES will focus its efforts on enhancing military satellite systems, he added.
A potential customer is Loral Space & Communications, which is considering the HYPRES solution for its next generation of satellites, not for the cryogenics, but rather for the solution’s advantages concerning reliability, gain and power consumption, Dunnegan said.
“Out in space, you don’t have to deal with the temperature,” Dunnegan said. “We’re running this under cryogenical cooling, but out there, it’s not going to be a big factor. This has piqued a lot of people’s interest.”